skills.rest
CarbonTails
CarbonTailsOFFICIAL
Explore all skills by this author →
GitHub

skill-creator

Create, modify, and optimize AI skills with test setup and feedback iteration.

Updated Jun 15, 2026
Search Tags:#skill development#AI optimization#skill iteration#performance measurement#AI programming
One-click install
npx skills add https://github.com/CarbonTails/Di-Vincis-Atelier --skill skill-creator-carbontails
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/CarbonTails/Di-Vincis-Atelier/tree/main/.cursor/skills/skill-creator
Command: npx skills add https://github.com/CarbonTails/Di-Vincis-Atelier --skill skill-creator-carbontails

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires subagents, grading, comparison, scripts, references, assets, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

The Skill Unit addresses the challenges of skill development, including creating new skills, modifying existing ones, and measuring their performance. It empowers users to iterate on their skills and improve their triggering accuracy.

Core Features & Use Cases

  • Skill Creation: Guide users through creating new skills from scratch, including capturing intent, writing drafts, and setting up test cases.
  • Skill Modification: Assist in editing and optimizing existing skills based on performance feedback.
  • Performance Measurement: Run evaluations to test skill performance, perform variance analysis, and optimize skill descriptions for better triggering accuracy.
  • Use Case: A user with limited programming knowledge wants to create a skill for processing customer support tickets. The skill creator guides them through each step, from defining the skill's purpose to running tests and improving the skill based on feedback.

Quick Start

Use the skill-creator skill to create a new skill for processing customer support tickets. Follow the prompts to define the skill's purpose, write a draft, set up test cases, and iterate on the skill based on feedback.

Frequently Asked Questions about skill-creator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create and test AI skills from scratch?▼

Skill creation involves defining intent, writing drafts, setting up test cases, and iterating based on feedback. This process guides you through each step to ensure proper skill development and accurate triggering.

What is blind comparison in skill iteration?▼

Blind comparison is a rigorous testing mechanism for skill iteration that evaluates performance objectively. It uses subagents for parallel execution to compare and grade different skill versions without bias.

Can I optimize skill descriptions for better triggering accuracy?▼

Yes, you can optimize skill descriptions for better triggering accuracy. The system runs evaluations, performs variance analysis, and provides feedback to refine descriptions so the skill activates correctly in the right contexts.

Do I need programming knowledge to develop AI skills?▼

Limited programming knowledge is sufficient to develop AI skills. The system guides you through defining purposes, drafting, and testing, making skill development accessible while supporting advanced optimization.

How does performance measurement work for AI skills?▼

Performance measurement for AI skills runs evaluations to test functionality, performs variance analysis, and grades results using subagents. This identifies weaknesses and guides iterative improvements for better triggering accuracy.

What are the limitations of using subagents for skill testing?▼

Using subagents for skill testing requires proper setup for parallel execution, grading, and comparison. Without these dependencies configured, the system cannot perform the blind comparisons needed for rigorous performance measurement.

Related Skills

Explore More Software Engineering Skills→
eglenn-deveglenn-dev
1

skill-creator

Create, refine, and evaluate AI skills with SKILL.md metadata and test cases.

Community
Advanced
earlyseasonearlyseason

pinchbench

Evaluate OpenClaw agent performance on real-world tasks using Python scripts.

Community
Advanced
NodeJSmithNodeJSmith
1

mine.skill-eval

Automate evaluation and comparison of AI skill variants via structured grading and statistical analysis.

Community
Advanced
yasuwrldhyperyasuwrldhyper

skill-creator

Create and refine AI skills with SKILL.md metadata and benchmark testing.

Community
Advanced